Such a monitoring system has been described in my prior U.S. Pat. Nos. 3,737,010, 3,747,011, 3,747,012, 3,919,661, 3,932,774, 3,932,803 and 3,935,542. As particularly disclosed in U.S. Pat. No. 3,932,803, the contactless motion detector may comprise an oscillator and an associated trigger amplifier whose energizing circuit includes a storage capacitor which is charged via a two-conductor supply network from a source of pulsating direct current by way of a constant-current unit in parallel with the anode/cathode path of an output thyristor acting as a voltage-controlled variable-impedance means. In one embodiment of the system described and claimed in the latter patent, the output thyristor and the constant-current device lie in series with an ancillary thyristor which shunts the storage capacitor and is separated therefrom by a decoupling diode. When the output thyristor is triggered by a signal from the oscillator, the ancillary thyristor is fired through a Zener diode inserted between its anode and its gate; thus, the two series-connected thyristors conduct simultaneously in series with a load such as a relay whose operated (or unoperated) state indicates an abnormal condition, e.g. the fact that the oscillator has detected the approach of a metallic element.
Though the system just described operates generally satisfactorily, a drawback resides in the fact that the load is energized through a resistance representing the sum of the resistance of the two series-connected thyristors in their conductive state. To eliminate this drawback, I have disclosed in my first-mentioned copending application Ser. No. 782,400 an improved system in which the resistance in series with the load is still further reduced in its high-current phase to increase the sensitivity of the system.
According to that improvement there is provided a first electronic control device having a pair of main electrodes respectively tied to the leads of the supply network and further having an input electrode, or gate, connected to one of its main electrodes (i.e. its anode if that control device is a thyristor or similar binary electronic switch such as a triac) through an operating circuit which includes a second electronic control device with input connections to the detector for changing the conductivity of the first control device in response to variations in the output signal of that detector, thereby modifying the energization of the load. Thus, the current flow through the load in a state of high conductivity of the first control device is limited practically exclusively by the resistance of that one device (aside from the internal resistance of the current source which preferably, as in the systems of my prior patents, includes a full-wave rectifier). Instead of binary electronic switches such as thyristors I may also use analog-type current regulators such as transistors establishing an entire gamut of energization rates in lieu of just a switchover between "on" and "off" states. Furthermore, the active component of the detector need not be an oscillator but could be an impedance bridge, a field plate or any of a variety of electromagnetic, photoelectric or other transducers responding to a predetermined change of an external condition to be monitored.
The storage capacitor insuring the continued energization of the detector, regardless of the state of conductivity of the electronic control devices, is connected across the leads of the supply network in series with a current-limiting device, more specifically a constant-current unit as in my prior U.S. Pat. No. 3,932,803. A decoupling diode is inserted between the storage capacitor and the ancillary thyristor (or other electronic switch constituting the aforementioned second control device), triggerable by the detector, activating the main thyristor (or other electronic switch representing the first control device) whose conduction raises the load current to its maximum value. The operating circuit lying between the anode and the gate of the main thyristor advantageously includes an impedance, preferably an electronic breakdown device such as a Zener diode, delaying the firing of the main thyristor after the ancillary thyristor has been triggered. During this brief delay period, the conduction of the ancillary thyristor recharges the storage capacitor through the decoupling diode while the constant-current unit is practically short-circuited by that conduction. A stabilizing resistor inserted between the gate and the cathode of the main thyristor lies in series with the aforedescribed operating circuit to limit the flow of gate current in that thyristor. The firing of the main thyristor virtually short-circuits the series combination of ancillary thyristor, Zener diode and stabilizing resistor so that the latter thyristor ceases to conduct. With pulsating or raw-rectified current supplied by the source, the main thyristor is also cut off at the end of each pulsation during which it has been rendered conductive; this results in a periodic recharging of the storage capacitor at the beginning of each new half-cycle (or full cycle in the case of half-wave rectification).
In the system just referred to, as well as in those of my prior patents wherein a storage capacitor shunted by a charge-limiting electronic breakdown device such as a Zener diode is charged through a constant-current unit, the breakdown device draws practically the full output current of the constant-current unit upon becoming conductive when the storage capacitor reaches a predetermined threshold voltage. This current flow is generally considerably greater than would be necessary in order to maintain conduction of the breakdown device until the capacitor charge has decayed below the sustaining level. It therefore constitutes a wasteful expenditure of energy and may also lead to objectionable heating of the load which lies effectively in series with the breakdown device. In a binary switching system, in which the load is to be in a state of minimum energization when the main or output thyristor is cut off, the existence of a large load current in the quiescent state is especially undesirable.
My other copending application referred to above, Ser. No. 786,805, describes and claims a further improvement through which the sustaining current drawn by a conductive charge-limiting electronic breakdown device is minimized. According to this latter improvement, a system of the general type described and claimed in my prior U.S. Pat. No. 3,932,803 is modified by connecting an electronic breakdown device such as a Zener diode between one of the output leads of the direct-current source, i.e. a lead which has an extension included in the energizing circuit for the motion detector, and a control input of a constant-current unit normally supplying charging current to a storage capacitor in series therewith, this Zener diode being nonconductive as long as the charging voltage of the capacitor lies below a predetermined level. When that level is reached, the voltage across the Zener diode equals its breakdown threshold whereupon that diode conducts and establishes a reference potential which, together with the rising capacitor voltage, throttles -- possibly down to complete cutoff -- the flow of charging current into the capacitor while drawing a reduced sustaining current from a section of the constant-current unit which lies between that control unit and the other lead of the d-c source.
More particularly, and as also disclosed in my prior U.S. Pat. No. 3,932,803, a constant-current unit for the charging of a storage capacitor in the presence of widely varying supply voltages (ranging, for example, between 40 and 250 V) comprises two parallel impedance branches, one of these branches including a transistor serving as an analog-type current regulator while the other branch comprises a voltage divider with a tap tied to the base of this current-regulating transistor. In the improved system, this tap constitutes the aforementioned control input of the constant-current unit whereby the sustaining current for the charge-limiting Zener diode is drawn only through one or more resistors forming part of the voltage divider; with the emitter of this transistor connected to the storage capacitor (usually through a resistance), this emitter-follower connection progressively reduces the conductivity of the transistor once the Zener threshold has been exceeded.
Even with the improvements first disclosed in my aforementioned copending applications, however, the load in its quiescent state is still traversed by a significant current drawn by a pilot transistor, or equivalent electronic switch, which is cut off by the oscillator only when the variable-impedance means in series with the source and the load is to be made highly conductive by the triggering of a thyristor of the like forming part thereof. In my prior systems, including that of U.S. Pat. No. 3,932,803, this pilot transistor is shown connected across the two supply leads in series with a resistor which limits but obviously does not cut off the current flow therethrough in the quiescent state.